Forest smoke and fire detection method based on target detection, storage medium and equipment

Abstract

The invention discloses a forest smoke and fire detection method based on target detection, a storage medium and equipment, and belongs to the field of computer vision target detection and forest smoke and fire video monitoring. The method includes: acquiring a forest smoke and fire image; performing forward calculation on the forest smoke and fire image by adopting a neural network model comprising a multi-level network to obtain a smoke and fire area and a confidence coefficient thereof; comparing the confidence coefficient with a confidence coefficient threshold value, and when the confidence coefficient is larger than the confidence coefficient threshold value, considering that a fire occurs, and when the confidence coefficient is smaller than or equal to the confidence coefficient threshold value, considering that no fire occurs. According to the invention, the network model with high precision is obtained by using small samples, the influence of the environment is small, the detection accuracy is high, false alarm is not easy to occur, and the bounding box of the smoke and fire area can be obtained, so that the fire occurrence position can be judged, and firefighters can quickly arrive at the fire area to extinguish the fire. Meanwhile, the disclosed storage medium and equipment can be directly deployed and used for forest smoke and fire detection.

Description

technical field [0001] The invention belongs to the fields of computer vision target detection and forest firework video monitoring, and in particular relates to a forest firework detection method, storage medium and equipment based on target detection. Background technique [0002] Forest fires are a common natural disaster. The casualties and property losses caused by forest fires in the world are huge every year. For example, the recent California forest fires, Indonesian forest fires, and Amazon forest fires have not only caused a large number of Casualties and property losses have also caused serious impacts on the natural environment. In the fire fighting process, a lot of manpower and material resources have been invested, but it is still difficult to effectively extinguish forest fires. Forest fires have caused serious threats to the survival of humans and animals. my country is a country with many forests. Every year, tens of thousands of casualties and hundreds of...

AI Technical Summary

Problems solved by technology

And when the method of image classification is used, the specific location where the fireworks occur cannot be accurately detected. Therefore, when a fire breaks out in the forest, it is difficult for firefighters to quickly reach the designated location to fight the forest fire.

[0005] In summary, the existing forest fire detection methods are easily affected by the external environment, the accuracy of detection is low, and they cannot accurately detect the location of forest fire, which affects forest fire fighting.

Images